ZOLL Medical Corporation
269 Mill Road
Chelmsford, MA 01824
Phone 978-421-9655
Fax 978-421-0025
www.zoll.com
8
Impact of Manual CPR on
Increasing Coronary Perfusion Pressure
In sudden cardiac arrest cases, the ability to adequately perfuse the brain and heart
during resuscitation is of critical importance. The problem is that manual chest
compressions during CPR provide only one third of normal blood supply to the
brain.1 Even more troubling, manual CPR provides only 10% to 20% of normal
blood flow to the heart.1 In fact, the role of CPR to provide flow to the heart muscle
may have historically been under-emphasized. While defibrillation is the definitive
therapy for ventricular fibrillation (VF), its efficacy is severely limited by response
time, and its success is also dependent on circulation.2 Quite literally, it may be
impossible to restart a heart after several minutes without first providing adequate
coronary blood flow.2
This paper discusses defibrillation and
manual CPR, explores the issues relating
to coronary blood flow, and describes a new
approach to perfusing the heart and brain during
cardiac arrest.
Time Limitations of Defibrillation
When defibrillation is indicated for sudden cardiac
arrest caused by VF, time to treatment is critical to
its success. In a study of cardiac arrests in casinos
that had automatic external defibrillators (AEDs)
applied, a 74% survival rate was observed in patients
receiving a first shock within 3 minutes of arrest.3
If the first shock was not delivered within 3 minutes,
the survival rate dropped to 49%, suggesting that
even in an ideal setting where arrests are witnessed
and defibrillators are readily available, time is a
critical factor in determining survival.
According to the study:
Intervals of no more than 3 minutes from
collapse to defibrillation are necessary to
achieve the highest survival rates.3
But in the EMS setting, a 3-minute interval from
arrest to first shock is generally not possible, and
EMS agencies rarely achieve the high survival rates
seen in the casino study.
To address this reality, many communities
have implemented EMT-D programs and
Public Access Defibrillation (PAD) programs.
These programs are important and need to
be supported, but they are unable to provide
early defibrillation to the majority of patients
who need it.
Over 70% of cardiac arrests outside the hospital
occur at home where no AED is available, and
therefore the vast majority of cardiac arrest patients
do not receive defibrillation within the crucial first
3 minutes.4
Another limitation is that defibrillation is not
indicated in at least half of all cardiac arrest patients.
That’s because more than 50% of patients in cardiac
arrest do not exhibit VF when the rescuer arrives.5
So, while defibrillation is the definitive treatment
for VF, countershock alone does not ensure
survival. Recognition of these facts has led to
a search for additional strategies to improve
myocardial perfusion as a means of increasing
survival in cardiac arrest.
The Important Role Circulation
Plays in Defibrillation
A growing body of evidence suggests that beyond
a delay of about 3 minutes, reestablishing
blood flow before defibrillation may improve
the efficacy of electrical countershock.
In fact, animal studies have demonstrated an
increased survival when, after several minutes
of arrest, subjects receive a brief round of CPR
before defibrillation.1
1
When a Seattle EMT-D program did not produce
the anticipated improvement in human survival,
the protocol was modified to include CPR
before defibrillation.
Though not a prospective, randomized trial, this
study of more than 1000 patients does speak to the
crucial role of circulation, as provided by rescuer
CPR, in increasing survival.
A recently reported prospective study followed
a similar protocol where patients were randomly
treated with either 3 minutes of CPR first or with
immediate countershock. This prospective study
showed that for patients treated after a response
interval of 5 or more minutes, a significantly higher
number had Return of Spontaneous Circulation
(ROSC), survival to discharge and one-year survival
when CPR was provided before defibrillation. See
Table 1. In this patient group, 58% experienced
ROSC vs. 38% of the control group, and the one-year
survival was 20% vs. 4% for the control group.6
Figure 1. % Survival According to Response Intervals
(minutes) of the First Arriving Unit2
The revised protocol directed the provision
of approximately 90 seconds of CPR before
defibrillation for patients in cardiac arrest. Under
this protocol, for patients whose initial presenting
rhythm was VF, the number of neurologically intact
survivors increased to 23% when compared with
a historical control of 17% neurologically intact
survivors (p=0.01).2 This result includes both
witnessed and unwitnessed arrests.
The efficacy of administering defibrillation
first appears to diminish over time and, after
approximately 3 minutes, better results are
achieved by delivering CPR prior to defibrillation.
The improvement in overall survival was greatest
in patients with a response interval of 4 minutes
or more after arrest (Figure 1). Survival for
these patients as a group increased from 17%
to 27% (p=0.01). 2
Both the Cobb and Wik studies highlight
the importance of circulation for increasing
survival. The question arises as to how well CPR
establishes circulation.
Myocardial and Cerebral
Perfusion in CPR
Even when performed by experts, manual CPR
produces only about 30% of blood flow to the
brain and a meager 10% to 20% of normal blood
flow to the heart.1 Due to its unique physiological
characteristics the heart is more difficult to perfuse.
The Mechanism of Heart Perfusion
When beating spontaneously, the heart is perfused in
the relaxation phase (diastole) that occurs between
contractions. As the heart contracts (systole), blood
is ejected out of the left ventricle, past the aortic
valve, and into the aorta. At this stage, though, the
blood doesn’t flow freely into the coronary arteries
or the myocardium. The reason: The force of the
heart’s contraction is greater than the force driving
blood into the coronary arteries. It is only with the
Table 1. Survival For Patients With a Response Interval of 5 or more minutes6
ROSC
2
Survival to Discharge
1-Year Survival
Group “A”:
3 minutes of CPR
37 (58%)
14 (22%)
13 (20%)
Group “B”:
Defibrillation First
21 (38%)
2 (4%)
2 (4%)
p<0.03
p<0.003
p<0.003
return of diastole that blood flows from the aorta into
the coronary arteries, perfusing the heart muscle.
Increased CPP Correlates
with Survival and ROSC
During CPR, the relaxation phase of chest compressions is similar to diastole, and maintaining the
proper balance between compression and relaxation
ensures adequate blood flow to the heart. This is
the foundation for the American Heart Association’s
recommended CPR duty cycle of 50% (the ratio of
time spent under compression to time spent in the
relaxation phase). 7
The clearest link between CPP and the likelihood
of a return of spontaneous circulation (ROSC)
is documented by Paradis et al. 10 CPP was
measured in 100 Emergency Department (ED
cardiac arrest patients. A definite correlation was
noted between peak CPP and ROSC (Figure 2).
How do we assess the adequacy of blood flow to
the heart during CPR? Commonly, the effectiveness
of manual CPR is assessed by checking for a pulse
generated by chest compression. The presence of
a pulse is a positive sign, especially when it can be
palpated with each compression. But it does not
necessarily mean that blood flow to the heart is
adequate. 8
According to the American Heart Association:
No studies have shown the clinical utility
of checking pulses during ongoing CPR.
…The important pressure for perfusion of the
myocardium is coronary perfusion pressure.8
The Importance of
Coronary Perfusion Pressure
Coronary perfusion pressure (CPP) is an indicator of
coronary flow. When CPP increases, so does blood
flow to the myocardium.9
CPP is the difference between the aortic pressure and
the right atrial pressure during diastole expressed in
millimeters of mercury (mm Hg).
CPP (mm Hg) = AP (mm Hg) – raP (mm Hg)
Aortic pressure (AP) is the driving force behind
coronary blood flow, but blood flow to the
myocardium is resisted by the pressure in the
coronary venous system. Therefore, the driving force
for coronary blood flow (aortic pressure) less the
pressure resisting flow (right atrial pressure) yields
the blood pressure gradient for that vascular bed,
and blood flow is related to this pressure gradient.
Measurement of CPP is an invasive research
technique and is not routinely available or practical
in the CPR setting. However, the desirability of
increasing CPP has tremendous clinical significance
and is a useful tool in clinical research.
Figure 2. CPP and ROSC in 100 ED Cardiac Arrest Patients10
Eleven (79%) of the 14 patients with a CPP greater
than 25mm Hg had a return of spontaneous
circulation, while no patient with a peak CPP of
less than 15mm Hg experienced such a return.
According to the study:
Return of spontaneous circulation and
survival from an arrest have been clearly
linked to the ability to achieve a CPP greater
than 15mm Hg.10
The problem is the difficulty in achieving and
maintaining CPP above 15mm Hg through
conventional CPR. In the 100 patients studied by
Paradis, conventional CPR provided a mean CPP
of only 12.5mm Hg, indicating that conventional
CPR cannot reliably provide the CPP necessary for
adequate ROSC and survival. As the result of this
information, efforts to develop new methods for
increasing myocardial perfusion during CPR have
been sought.
3
Rescuer CPR Performance
As discussed before, CPR typically does not
perfuse the heart or brain well under the best of
circumstances. In addition, there are inconsistencies
inherent in manual CPR.
Several facts emerge from the literature evaluating
CPR performance.11,12
• Rescuers have difficulty accurately
determining the correct depth of
compression.
• Rescuer fatigue sets in within 1 minute of
CPR, measurably affecting the quality of
chest compressions.
• Rescuers cannot accurately perceive their
own fatigue.
These shortcomings are not related to individual
rescuers, but rather to the complex and demanding
nature of manual CPR.
In 1998 Ochoa reported on a study of 38 hospital
clinicians which found that in the second minute
of chest compressions, only 24.9% were done
correctly.11 Furthermore, most rescuers did not
perceive any fatigue until after 3 minutes. And 26%
of those studied did not perceive any fatigue after 5
minutes, even though a decrease in performance was
observed after only 1 minute. Based on this study
and a similar study of EMS rescuers by Hightower,12
it would seem vital to rotate rescuers regularly.
In an attempt to create a useful guideline, another
study evaluated teams of two and three rescuers
performing compressions, rotating after periods of
1, 2 or 3 minutes. The researchers concluded that
to maintain technically correct performance, chest
compressions should be performed over periods
of 1 minute with at least three rescuers rotating
every 1 minute.13 Unfortunately, this scheme still
results in less than optimal CPR, not to mention the
impracticality of administering CPR in this way.
Effects of Pausing Manual CPR
Frequent rotation of rescuers may improve
performance, but it introduces a new concern. When
compressions are stopped, even for a few seconds,
CPP drops significantly and ROSC becomes less
likely. 1
Figure 3. The AutoPulse™ Non-invasive Cardiac Support Pump
Physical Demands of Manual CPR
According to the AHA Guidelines 2000, the chest
must be compressed at a rate of 100 compressions
per minute, to a depth of 1½ to 2 inches, with proper
hand placement, and a 50% duty cycle should be
maintained (50% of time under compression, 50%
percent under relaxation).7 This is simply more
than most rescuers can physically achieve and
certainly more than anyone can perform beyond a
few minutes.
4
Earlier CPR guidelines called for a ratio of 5
compressions to 1 ventilation, while current AHA
guidelines specify a 15:2 ratio. One reason for this
change is that CPP is higher after 15 uninterrupted
chest compressions than after 5 compressions. 1
Any benefit gained from ventilating every fifth
compression is outweighed by the subsequent loss
of CPP. Additional evidence shows that pauses in
compressions decrease both CPP and the probability
of ROSC.1
In summary, data reveals that even the most
highly trained rescuers can rarely perform manual
CPR correctly for more than 1 minute, and
even when they can, they are unable to perfuse
the heart and brain sufficiently for recovery.
For these reasons, manufacturers have attempted
to develop mechanical devices that can avoid
rescuer fatigue while providing accurate, consistent
chest compressions.
The Ideal CPR Device
compressions to the patient’s chest.
In a recent article reviewing the literature on clinical
and laboratory use of external and noninvasive
mechanical CPR devices, the author states:
In addition, the AutoPulse has been specially
designed to be:
The goal must be to provide mechanized
equipment that is easy to apply and use on the
patient as early as possible. The device must
also produce a haemodynamic profile better
or at least as good as optimally performed
manual ECC.15
• Rapidly deployable in the field.
• Automatically adjustable to the patient.
• Practical for rescuers of all skill levels.
Several mechanical CPR devices have been
developed, but most of these have no known data
that show consistently improved CPP over properly
performed manual CPR.
Furthermore, most of the devices:15
• Have operational limitations due to
application time.
• Are cumbersome to install and operate.
• Are heavy and unstable on the chest.
They do not meet the criteria for either improved
hemodynamics or ease of use.
The AutoPulse™ Non-invasive
Cardiac Support Pump
The AutoPulse Non-invasive Cardiac Support Pump
from ZOLL Medical Corporation is a new device
that deploys in seconds to provide automated chest
compressions at a consistent rate and depth and
standard duty cycle during CPR. In 2001, the FDA
cleared the use of the AutoPulse as an adjunct to
manual CPR for commercial distribution.
The AutoPulse is a portable, automated chest
compression device intended for use as an adjunct to
manual cardiopulmonary resuscitation in the adult
atraumatic cardiac arrest population. The AutoPulse
offers a more efficient method of generating chest
compressions during CPR, and results from recent
studies provide evidence of the improvement in
blood flow when the AutoPulse is used compared
to conventional CPR.
The device consists of a single, integrated platform
that contains a microprocessor-based control
system, an electromechanical drive system, and
a user interface panel. A single-patient-use chest
compression assembly provides pre-programmed
Figure 4. Human Study CPP Result16,10
Efficacy of the
AutoPulse—Human Study
In a study presented at the NAEMSP 2003 Annual
Meeting, the AutoPulse demonstrated a significant
increase in CPP compared to the CPP generated from
aggressively performed manual CPR. Subsequent to
IRB approval, a total of 31 sequential subjects with
in-hospital sudden cardiac arrest were screened,
and 16 were enrolled. All subjects received prior
treatment for cardiac disease and most had systemic
co-morbidities. Following a minimum of 10 minutes
of failed ACLS and catheter placement, the intubated
and ventilated subjects received alternating manual
and AutoPulse chest compressions for 90 seconds
each.
The AutoPulse demonstrated a significant increase
in CPP compared to the CPP generated from
aggressively performed manual CPR. Specifically, the
AutoPulse was able to produce a mean CPP above the
previously described 15mm Hg threshold necessary
for ROSC which manual chest compressions did not
achieve.16
The mean CPP generated with AutoPulse
and with manual CPR treatments are shown in
Figure 3. For the 16 patients in the study the mean
5
CPP for AutoPulse was 33% higher than manual
CPR (20mm Hg vs. 15mm Hg, p < 0.05).
Efficacy of the
AutoPulse—Animal Study
An animal study was also conducted that
allowed actual blood flow measurements to be
performed in addition to CPP measurements.
This study, led by Henry Halperin M.D. at
Johns Hopkins University, School of Medicine, was
performed to assess hemodynamics with the use
of the AutoPulse device as compared to manual
CPR. In order to eliminate the effects of fatigue
and inconsistencies inherent in manual CPR, the
Thumper® system (Michigan Instruments) was used
to provide a consistent form of manual CPR.17
Ventricular fibrillation (VF) was induced in 10
pigs and after one minute, CPR was performed.
AutoPulse CPR (A-CPR) and conventional CPR
(C-CPR) were performed in random order.
For the Basic Life Support or “BLS” scenario, no
epinephrine was used. For the Advanced Life Support
or “ALS” scenario, a 0.5mg bolus of epinephrine was
administered followed by 0.004mg/kg/min infusion
of epinephrine. CPR was initiated simultaneously
with the administration of epinephrine.
CPP was measured as well as regional blood flow.
Regional flows were measured with neutronactivated microspheres.
The results (Table 2) provided very encouraging
evidence of the potential for improved hemodynamics with AutoPulse compared with
conventional CPR:
• Without the use of epinephrine (BLS), the
AutoPulse was able to produce a mean CPP
of 21mm Hg—well above the important
15mm Hg threshold necessary for ROSC
as previously described. In comparison,
the mean CPP was only 14mm Hg for
conventional CPR.
• The AutoPulse produced 36% of normal
coronary flow vs. only 13% produced
by conventional CPR without the use of
epinephrine (BLS).
• When epinephrine was administered to
the animals early in the course of the arrest
(ALS), the AutoPulse generated blood flow
to the heart and brain that were equivalent
to pre-arrest levels of flow.
Based on these data, the AutoPulse clearly meets the
criterion for demonstrating a hemodynamic profile
better or at least as good as optimally performed
manual CPR. And its easy operation and rapid
deployment in the EMS setting more than satisfies
the requirement for practical, efficient use in realworld situations.
Table 2. Hemodynamic Parameters Measured in a Porcine Model of Ventricular Fibrillation (n = 10)
ALS
BLS
Myocardial Flow
(% of Pre-Arrest
Level)*
Cerebral Flow
(% of Pre-Arrest
Level)*
CPP (mm Hg)*
AutoPulse
36%(±12%)
36%(±10%)
21(±2)
Conventional CPR*
13%(±3%)
28%(±11%)
14(±2)
p=0.07
p<0.6
p<0.001
AutoPulse with
epinephrine
127%(±36%)
129%(±27%)
45(±3)
Conventional CPR**
with epinephrine
29%(±11%)
31%(±6%)
17(±2)
p<0.02
p<0.003
p<0.001
*Mean ±S.E.
**For purposes of the study, the Thumper was used to provide consistent manual CPR
to eliminate the effects of fatigue and inconsistency inherent in human CPR.
6
Summary
References
Given the necessity of perfusing the heart well and
the difficulty of performing adequate manual CPR,
many have sought to develop a mechanical adjunct
capable of performing chest compressions. The
ideal device, according to experts, must improve
hemodynamics and be easy to use.
1.
Kern KB, Coronary perfusion pressure during cardiopulmonary
resuscitation. Bailliere’s Clinical Anaesthesiology. 2000; 14(3): 591-609.
2.
Cobb LA, Fahrenbruch CE, Walsh TR, Copass MK, et al. Influence of
CPR prior to defibrillation in patients with out-of-hospital VF. JAMA
1999; 281: 1182-1188.
3.
Valenzuela TD, Roe DJ, Nichol G, Clark LL, et al.. Outcomes of rapid
defibrillation by security officers after cardiac arrest in casinos. NEJM
2000; 343: 1206-1209.
4.
Engdahl J, Holmberg M, Karlson BW, et al. The epidemiology of out-ofhospital ‘sudden’ cardiac arrest. Resuscitation. 2002; 52(3): 235-245.
5.
Cobb L, Fahrenbruch C, Olsufka M, et al. Changing Incidence of Outof-Hospital Ventricular Fibrillation, 1980-2000. JAMA. 2002; 288(23):
3008-3013.
6.
Wik L, Hansen TB, Fylling F, et al. Delaying defibrillation to give basic
cardiopulmonary resuscitation to patients with out-of-hospital venticular
fibrillation. JAMA. 2003; 289(11):1389-1395.
7.
American Heart Association in collaboration with International Liaison
Committee on Resuscitation. Guidelines 2000 for Cardiopulmonary
Resuscitation and Emergency Cardiovascular Care: International
Consensus on Science, Part 3: Adult Basic Life Support. Circulation 2000;
102(suppl I): I-43.
8.
American Heart Association in collaboration with International Liaison
Committee on Resuscitation. Guidelines 2000 for Cardiopulmonary
Resuscitation and Emergency Cardiovascular Care: International
Consensus on Science, Part 3: Adult Basic Life Support. Circulation 2000;
102(suppl I): I-108.
9.
Halperin HR, Tsitlik JE, Guerci AD, et.al. Determinants of blood flow to
vital organs during cardiopulmonary resuscitation in dogs. Circulation.
1986; 73(3): 539-550.
The portable, easy-to-deploy AutoPulse outperforms
manual CPR in generating blood circulation to all
organs—including the heart. It has been shown to
increase CPP in both animals and humans over and
above optimally performed manual CPR. Further,
it has been shown in animals to produce levels of
blood flow that are greater than pre-arrest levels of
flow with the use of epinephrine.
10. Paradis NA, Martin GB, Rivers EP, Goettling MG, et al. Coronary
perfusion pressure and the return of spontaneous circulation in human
cardiopulmonary resuscitation. JAMA 1990; 263: 1106-1113.
11. Ochoa FJ, Ramalle-Gómara E, Lisa V, Saralegui I. The effect of rescuer
fatigue on the quality of chest compressions. Resuscitation 1998; 37:
149-152.
12. Hightower D. Decay in quality of closed-chest compressions over time.
Ann Emerg Med 1995; 26: 300-303.
13. Huseyin TS, Matthews AJ, Wills P, O’Neill VM. Improving the
effectiveness of continuous closed chest compressions: an exploratory
study. Resuscitation 2002; 54: 57-62.
14. American Heart Association in collaboration with International Liaison
Committee on Resuscitation. Guidelines 2000 for Cardiopulmonary
Resuscitation and Emergency Cardiovascular Care: International
Consensus on Science, Part 3: Adult Basic Life Support. Circulation 2000;
102(suppl I): I-41.
15. Wik L. Automatic and manual mechanical external chest compression
devices for CPR. Resuscitation 2000; 47: 7-25.
16. Timmerman S, Cardoso LF, Ramires JA, et al. Improved hemodynamics
with a novel chest compression device during treatment of in-hospital
cardiac arrest. Prehospital Emergency Care. 2003; 7(1); 162.
17. Halperin HR, Paradis N, Ornato JP et al. Improved hemodynamics with a
novel chest compression device during a porcine model of cardiac arrest.
Circulation 2002: 106 (19) (Suppl II): 538.
7